MTD detector using convolutional neural networks

A detector based on joint time-frequency signal analysis and convolutional neural networks is proposed for radar detection in highly complex and nonstationary cluttered environments. This detector is coherent and monocell, i.e. it works with the complex envelope of the echoes from the same range cel...

Full description

Saved in:
Bibliographic Details
Main Authors: Grajal, J., Quintas, A.G., Lopez-Risueno, G.
Format: Conference Proceeding
Language:English
Subjects:
Cellular neural networks
Clutter
Convolution
Degradation
Envelope detectors
Neural networks
Radar detection
Signal analysis
Spaceborne radar
Time frequency analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A detector based on joint time-frequency signal analysis and convolutional neural networks is proposed for radar detection in highly complex and nonstationary cluttered environments. This detector is coherent and monocell, i.e. it works with the complex envelope of the echoes from the same range cell, and exhibits joint CFAR and MTD characteristics. It includes a pre-processing time-frequency block which provides a constant false alarm rate (CFAR) behaviour regarding the clutter power when normalization is utilized. Multiple targets can be also resolved in the same resolution cell (MTD) if the neural network presents multiple outputs.
ISSN:1097-5659
2375-5318
DOI:10.1109/RADAR.2005.1435941